| Methods |
Study design: cluster‐randomized controlled trial
Study grouping: parallel group
Study aim: "to assess whether a participatory parent‐focused approach using parents as agents of behavioral change enhances the efficacy of a preschool physical activity intervention" (quote)
Study period: randomization: August 2008; data collection start date: September 2008; data collection end date: May 2010
Total number of arms: 2
Description of intervention arms: 1. Preschool physical activity program and participatory intervention (child + caregiver); 2. Preschool physical activity program (child only)
Number of clusters per arm: child + caregiver arm: n = 19; child‐only arm: n = 20
Average cluster size: 21.18 children
Sample size justification and outcome used: to detect a difference of 19.05 accelerometry counts per 15 seconds (0.5 standard deviation) between the child + caregiver arm and the child‐only arm, with 90% power and a 5% probability of a type 1 error, 504 children in 24 schools were required. Estimate of standard deviation (38.1 counts per 15 seconds) came from unpublished data from pilot studies with children aged 4 to 5 years. Study authors conservatively estimated an intracluster correlation (ICC) of 0.1 (design effect = 3). Recruitment aimed to enroll 280 children per arm (560 children total) to account for an anticipated 10% loss‐to‐follow‐up rate
Unit of allocation: preschools
Missing data handling: "in all models, missing data of outcomes were imputed by a cross‐sectional regression imputation approach" (quote). Age, gender, and measurement time were used as predictors in imputation analyses; 14.6% of anthropometry data and 31% of physical activity data were imputed using this approach. The pattern of missing physical activity monitoring data did not differ between the 2 arms. Missing data between children with and without physical activity monitoring differed only with respect to immigrant background. Sensitivity analyses were used to assess the effects of missing data
Reported limitations: 1. Children were sampled from preschools that voluntarily applied for a state‐sponsored physical activity program, which may have reduced generalizability; 2. Absence of children on measurement days may have introduced post‐randomization selection bias; 3. Difficult to attribute intervention effects to specific components due to site‐level adaptations; 4. The study did not include a control group that received no intervention; 5. Intervention facilitators were involved in the process evaluation, which might have biased these results
Randomization ratio and stratification: 1:1, stratified by school socioeconomic status (low, middle, high) and location (rural, non‐rural)
Participant compensation or incentive: not reported |
| Participants |
Baseline characteristics
Child + caregiver arm (intervention group)
Female (PROGRESS‐Plus): not reported Age in years (PROGRESS‐Plus): not reported Race/ethnicity/language/culture (PROGRESS‐Plus): not reported Place of residence (PROGRESS‐Plus): not reported Caregiver education (PROGRESS‐Plus): not reported Religion (PROGRESS‐Plus): not reported Household income/socioeconomic status (PROGRESS‐Plus): not reported Social capital (PROGRESS‐Plus): not reported Caregiver work hours and other characteristics that may indicate disadvantage (PROGRESS‐Plus): not reported Disability (PROGRESS‐Plus): not reported Sexual orientation (PROGRESS‐Plus): not reported Child weight status: not reported Child diet: not reported Child physical activity: MVPA (min/d), mean (SD): 37.31 (15.21); sedentary time (min/d), mean (SD): 631.3 (68.3) Caregiver weight status: not reported Caregiver diet: not reported Caregiver physical activity: not reported Caregiver civil status (PROGRESS‐Plus): not reported
Child‐only arm (control group)
Female (PROGRESS‐Plus): not reported Age in years (PROGRESS‐Plus): not reported Race/ethnicity/language/culture (PROGRESS‐Plus): not reported Place of residence (PROGRESS‐Plus): not reported Caregiver education (PROGRESS‐Plus): not reported Religion (PROGRESS‐Plus): not reported Household income/socioeconomic status (PROGRESS‐Plus): not reported Social capital (PROGRESS‐Plus): not reported Caregiver work hours and other characteristics that may indicate disadvantage (PROGRESS‐Plus): not reported Disability (PROGRESS‐Plus): not reported Sexual orientation (PROGRESS‐Plus): not reported Child weight status: not reported Child diet: not reported Child physical activity: MVPA (min/d), mean (SD): 36.42 (14.83); sedentary time (min/d), mean (SD): 631.4 (63.2) Caregiver weight status: not reported Caregiver diet: not reported Caregiver physical activity: not reported Caregiver civil status (PROGRESS‐Plus): not reported
Although not reporting by intervention arm, study authors do provide additional information on the overall study sample
Female (PROGRESS‐Plus), %: 48 Age in years (PROGRESS‐Plus): mean: 5.05 Race/ethnicity/language/culture (PROGRESS‐Plus): immigrant background, %: 37 Caregiver education (PROGRESS‐Plus): low, %: 25; middle, %: 55; high, %: 20 Child weight status: overweight, %: 5.4
Recruitment methods: all preschools that applied for the state‐sponsored physical activity program were approached about participating. Details on recruitment were not reported
Inclusion criteria: cluster: participation in an existing, state‐sponsored physical activity program; participants: children aged 4 to 6 years and caregivers
Exclusion criteria: cluster: fewer than 15 children participating in the state‐sponsored physical activity program or without an external physical activity teacher in the area; participants: not reported
Age of participating children at baseline: 4 to 6 years (preschool)
Total number randomized by relevant group: total across both study arms: n = 826; child + caregiver arm: n = 441; child‐only arm: n = 385
Baseline imbalances between relevant group: the child + caregiver arm had larger group sizes than the child‐only arm (mean cluster size 23 vs 19). No other significant differences in baseline characteristics were observed between study arms
Total number analyzed by relevant group: child + caregiver arm: n = 433; child‐only arm: n = 376
Attrition by relevant group: attrition rates were calculated as the number of children whose schools withdrew consent divided by the number of children randomized: child + caregiver arm: 1.8% (8/441), child‐only arm: 2.3% (9/385). Data for all other children were available or imputed
Description of sample for baseline characteristics reported above: total sample analyzed (n = 809) |
| Interventions |
Intervention characteristics
Child + caregiver arm (intervention group)
Brief name/description (TIDieR #1): state‐sponsored preschool physical activity program plus participatory caregiver‐focused intervention (Ene mene fit) Focus of intervention: physical activity Behavior change techniques: in addition to the child‐only intervention, the following techniques were applied separately or differently in the child + caregiver arm: "goals and planning," "feedback and monitoring," "social support," "shaping knowledge," "comparison of outcomes," "antecedents" Why: rationale, theory, or goal (TIDieR #2): caregivers and teachers have an important role in shaping children’s physical activity behaviors and opportunities. More sustainable preschool physical activity interventions that successfully involve caregivers and teachers are needed. As with the child‐only intervention, this intervention was based on general systems theory. Study authors also adopted a participatory approach in an effort to overcome the lack of choice and co‐determination that has hampered health promotion interventions that rely on educational and information‐based strategies How, where, and when and how much (TIDieR #6 to 8): in addition to the child‐only intervention, preschools received a participatory parent‐led intervention. After an introduction to the intervention at the school’s convocation of parents and teachers, 3 follow‐up parent‐teacher meetings were convened. Interested teachers and parents were encouraged to identify up to 4 projects in which all children in the preschool could take part. This intervention ran concurrently with the child‐only intervention for the first 6 months and then continued for an additional 3 months Who: providers (TIDieR #5): specially trained external physical activity teachers, caregivers, and preschool teachers Economic variables and resources required for replication: not reported Strategies to address disadvantage: not reported Subgroups: same as child‐only arm Assessment time points: baseline, 6 months (before end of intervention), 1 year (approximately 6 months after end of intervention) Co‐interventions: not reported What: materials and procedures (TIDieR #3 to 4): in addition to the child‐only intervention, the participatory parent‐focused intervention engaged caregivers in development and implementation of projects to promote physical activity. After an introduction to the intervention with a video and organized discussion, 3 follow‐up parent‐teacher meetings were convened: a workshop to select projects and develop project teams, team presentations to all parents to motivate participation, and a workshop for planning implementation. Sessions were facilitated and supported by external gym trainers. Caregivers and teachers received access to an intervention‐specific website and a printed book with 15 pilot‐tested project ideas. The physical activity projects established by caregivers and teachers were promoted by posted lists, and all children at the preschool could take part Tailoring (TIDieR #9): due to the participatory nature of the intervention, project components varied across sites. Parents and preschool teachers at all sites were provided with a list of 15 pilot‐tested project ideas. They were encouraged to identify up to 4 projects to pursue. Teams were encouraged to select project ideas from the list or to develop new ideas that were focused on promoting physical activity in a simple and sustainable way Modifications (TIDieR #10): not reported How well: planned and actual (TIDieR #11 to 12): a process evaluation following the RE‐AIM framework was used to assess the intervention. "In all, 33% of the eligible children and 46% of the parents were reached” (quote). Three preschools chose not to select and implement physical activity projects (adoption rate, n [%]: 15/18 [83]). Of the preschools that chose projects, 12 actually implemented project activities to some degree (Implementation rate, n [%]: 12/15 [80]). "The majority of projects (65%) were newly developed by parents, and many (44%) were transferred to the new school year" (quote) Sensitivity analyses: same as child‐only arm
Child‐only arm (control group)
Brief name/description (TIDieR #1): state‐sponsored preschool physical activity program Focus of intervention: physical activity Behavior change techniques: "shaping knowledge," "comparison of behavior," "repetition and substitution," "comparison of outcomes" Why: rationale, theory, or goal (TIDieR #2): children establish health behaviors early in life. However, few studies have examined physical activity interventions in preschools. In Germany, preschool attendance rates are high, but no uniform guidelines exist for the frequency of physical activity lessons in preschools. This intervention sought to encourage physical activity among preschool children. It was based on general systems theory, which suggests that behaviors are influenced by social networks, and behavior change efforts should involve multiple members or “agents” of an individual’s network (e.g. caregivers, teachers, siblings, peers) How, where, and when and how much (TIDieR #6 to 8): the state‐sponsored program lasted for 6 months and involved twice weekly 1‐hour gym classes (40 total) and 1 parent‐gym trainer meeting Who: providers (TIDieR #5): specially trained external physical activity teachers Economic variables and resources required for replication: not reported Strategies to address disadvantage: not reported Subgroups: in the published study protocol, study authors indicated planned subgroup analyses at the child and preschool levels. At the child level, they planned to compare outcomes for normal weight vs overweight children and for immigrant vs non‐immigrant children. At the school level, they planned to compare outcomes for rural vs non‐rural locations and based on intervention timing. We did not find reports of these analyses. Assessment time points: baseline, 6 months (before end of intervention), 1 year (approximately 6 months after end of intervention) Co‐interventions: not reported What: materials and procedures (TIDieR #3 to 4): gym classes were standardized. No information was provided on content or procedures Tailoring (TIDieR #9): not reported Modifications (TIDieR #10): not reported How well: planned and actual (TIDieR #11 to 12): not reported Sensitivity analyses: analyses were re‐run without imputed data to assess the effects of missing data. The direction of effects remained the same for all outcomes; however, only the reduction in sedentary behavior and the increase in general perceived health remained significant
|
| Outcomes |
The following instruments were used to measure outcomes relevant to this review at baseline, 6 months (end of intervention), and 1 year (follow‐up)
-
Children's dietary intake: parents were asked, "How many portions (size of a child's hand) of fruits/vegetables does your child eat on average per day?" and completed an adapted food frequency questionnaire that addressed sugar‐sweetened beverage consumption among other things
-
Children's physical activity levels and sedentary behavior: 1‐dimensional accelerometry in the vertical plane using Actiheart accelerometers (CamNtech, Cambridge, UK) and questions answered by parents from a validated survey instrument; for accelerometry data, only data from waking periods between 7:00 am and 9:00 pm from children with at least 1 weekday and 1 weekend day over a consecutive 6‐day period were included
-
Children's anthropometry: height measured with Seca Deutschland (Hamburg, Germany) and weight measured with Soehnle Pharo (Nassau, Germany)
BMI values were reported for end‐of‐intervention and follow‐up assessments Prevalence of overweight or obesity was not reported for end‐of‐intervention or follow‐up assessments and could not be retrieved from the study authors
The study authors note that 6‐month data were collected "shortly before the end of intervention" |
| Identification |
Study name: Ene mene fit
Country: Germany
Setting: preschools in Baden‐Württemberg, Germany
Types of reports: published protocol; published journal article; trial registration
Comments: used the following reports: (1) De Bock 2010, (2) De Bock 2013b, (3) Trial registration (Clinicaltrials.gov NCT00987532)
Author's name: Freia De Bock
Email: freia.debock@medma.uni‐heidelberg.de
Conflicts of interest: "the authors declare that they have no competing interests" (quote)
Sponsorship source: Landesstiftung Baden‐Wurttemberg |
| Notes |
|
| Risk of bias |
| Bias |
Authors' judgement |
Support for judgement |
| Random sequence generation |
Unclear risk |
Quote: "preschools were randomized (1:1) after stratification for aggregate SES [socioeconomic status] and geographic location (urban vs rural) using sealed opaque envelopes" (De Bock 2013b, p 65)
Judgment comment: the method used to generate the random sequence was not reported |
| Allocation concealment |
Low risk |
Quote: "preschool assignment was blinded through the use of sequentially numbered, sealed envelopes" (De Bock 2010, p 53) |
| Blinding of participants and personnel
All outcomes |
Low risk |
Quote: "due to the participatory nature of the study, it was not possible to blind participants and study personnel" (De Bock 2010, p 53 to 54)
Judgement comment: given that children were 4 to 6 years of age, their performance was not likely to have been influenced by lack of blinding |
| Blinding of outcome assessment
All outcomes |
Low risk |
Quote: "outcome assessors were blinded to group allocation as suggested for prospective randomized open trials with blinded evaluation" (De Bock 2013b, p 65)
Judgment comment: height and weight were measured and physical activity was assessed by accelerometer |
| Incomplete outcome data
All outcomes |
High risk |
Judgment comment: we calculated attrition rates as the number of children whose schools withdrew consent divided by the number of children randomized. Total attrition in the 2 relevant study arms was 2.1% (17/826). Differential attrition was 0.5% (8/441 vs 9/385). Additionally, 3 schools in the child + caregiver arm did not receive the intervention because the teachers disliked the measurements. However, the study authors did not report overall how many participants were lost to follow‐up or withdrew from the study. Missing data (31% of physical activity data and 14.6% of anthropometry data) were imputed for analyses. Sensitivity analyses to assess the effects of missing data on outcomes found that the direction of effects remained the same for all outcomes, but only the reduction in sedentary behavior and the increase in general perceived health remained significant |
| Selective reporting |
Low risk |
Judgment comment: the trial was registered and the protocol was published. Expected outcomes were pre‐specified and were addressed in the article |
| Recruitment bias |
Unclear risk |
Judgment comment: it is unclear whether children were recruited before or after randomization |
| Baseline imbalance |
Unclear risk |
Quote: "except for larger group sizes in the intervention arm (mean cluster size 23 (15–50) vs 19 (9–46)), there were no differences in baseline characteristics between the study arms" (De Bock 2013b, p 67)
Judgment comment: study authors did not report on similarities and differences between clusters |
| Loss of clusters |
High risk |
Judgment comment: study authors reported that 1 preschool in each study arm withdrew consent after randomization. A further 3 schools in the child + caregiver arm did not receive the intervention because teachers disliked the measurements |
| Incorrect analysis |
Low risk |
Quote: "the core model assumed a linear change of the outcomes with time and included two normally distributed random effects (one at the preschool level and one at the child level) to adjust for clustering in the data due to the hierarchic sampling scheme" (De Bock 2013b, p 67) |
| Comparability with individually randomized trials |
Unclear risk |
Judgment comment: information was insufficient to permit judgment |
| Other sources of bias |
Low risk |
Judgment comment: study authors reported that "absence of children on measurement days may have introduced post‐randomization selection bias" (quote; De Bock 2013b, p 72). Children without physical activity measurements were more likely to have an immigrant background. However, "the possible oversampling of a low‐risk population (without immigrant background) would rather lead to an underestimation of effect size, and is very unlikely to change the direction of effect" (quote; De Bock 2013b, p 72) |
